Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2502.20235 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866913710847557632 |
|---|---|
| author | Zhou, Yang Gao, Xu Chen, Zichong Huang, Hui |
| author_facet | Zhou, Yang Gao, Xu Chen, Zichong Huang, Hui |
| contents | Recent advances in generative diffusion models have shown a notable inherent understanding of image style and semantics. In this paper, we leverage the self-attention features from pretrained diffusion networks to transfer the visual characteristics from a reference to generated images. Unlike previous work that uses these features as plug-and-play attributes, we propose a novel attention distillation loss calculated between the ideal and current stylization results, based on which we optimize the synthesized image via backpropagation in latent space. Next, we propose an improved Classifier Guidance that integrates attention distillation loss into the denoising sampling process, further accelerating the synthesis and enabling a broad range of image generation applications. Extensive experiments have demonstrated the extraordinary performance of our approach in transferring the examples' style, appearance, and texture to new images in synthesis. Code is available at https://github.com/xugao97/AttentionDistillation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_20235 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Attention Distillation: A Unified Approach to Visual Characteristics Transfer Zhou, Yang Gao, Xu Chen, Zichong Huang, Hui Computer Vision and Pattern Recognition Recent advances in generative diffusion models have shown a notable inherent understanding of image style and semantics. In this paper, we leverage the self-attention features from pretrained diffusion networks to transfer the visual characteristics from a reference to generated images. Unlike previous work that uses these features as plug-and-play attributes, we propose a novel attention distillation loss calculated between the ideal and current stylization results, based on which we optimize the synthesized image via backpropagation in latent space. Next, we propose an improved Classifier Guidance that integrates attention distillation loss into the denoising sampling process, further accelerating the synthesis and enabling a broad range of image generation applications. Extensive experiments have demonstrated the extraordinary performance of our approach in transferring the examples' style, appearance, and texture to new images in synthesis. Code is available at https://github.com/xugao97/AttentionDistillation. |
| title | Attention Distillation: A Unified Approach to Visual Characteristics Transfer |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.20235 |